Questionnaire about the needs of Greek typography end-users

During the last few decades Greek typography has been strongly influenced by Latin elements. Often, there is substitution of Greek letters by Latin ones. A few designers try to preserve characteristic Greek typographic elements into their work, but overall Greek typography appears to trend towards Latin assimilation.

a) To define what constitutes Greek characteristics in a font,
b) To delineate the needs and problems faced by contemporary Greek typography end-users (graphic designers, book publishers of Greek texts, etc.)

a) How do you define or how do you approach the issue of a Greek characteristic in a font?
b) What questions would you have for end-users of your fonts?
c) Why do you design Greek fonts?

P.S. The results of this study, including all information from font-designers and end-users, will be posted in this blog in the end of June.

Tatiana Marza's picture

Thank you for posting the article of Yannis Haralambous.

I never got to thank you for the 'An Essay on Typography' of Eric Gill. Thank you!


Dear Gentlemen,

Writing the thesis on Latinization of Greek typography I faced a pragmatic question: why is so harmful this phenomena?
From the available bibliography, I could deduce that

1) every writing system is a cultural heritage and the identity of every nation
2) Latinization may demote cultural authenticity of a non - latin script and
3) reduce functionality

I cannot find a strong argument against Latinization.

I presume almost everyone would agree on three mentioned points, although the third, about functionality it is not quite clear to me. If it implies readability, at least for Greek text, I still do not see an extended problem. Recently I saw a short passage in a journal, where among other latinized greek letters 'η' was actually a latin 'n', no descending part. I read the text easily and observed the substitution only because I am aware of it. But there was nothing tragic in the situation.

Even if Latin alphabet will assimilate completely the Greek one, there always will exist a 'community' - editors of classical Greek and ecclesiastical texts which will use fonts related to Byzantine scripts.

From the discussions of this thread, although with an imprecise way, I understood that it is possible to preserve the characteristics of Greek letters and to give them a more modern form. But, it seems to me that Greek typography is 'trapped' into Byzantine/Renaissance/Baroque periods. While Latin letters give a designer more room for changes towards progress, Greek typography is far away from that. Do you think is possible to preserve Greek characteristics and to design a font which would resemble to the clarity and geometry of Latin one?

Greek nation is a contradictory one. It is steadily with one leg into their past and with other travelling through the entire world. But the same could be said about Arab countries. It is wrong or right, I don't know. But it seems to me that this is one side of human evolution - to tend to something new, something different.

Do you think only type designers are concerned with the issue of Latinization, while Greek society is not interested at all?

quadibloc's picture

@Tatiana Marza:
Do you think only type designers are concerned with the issue of Latinization, while Greek society is not interested at all?

I can't speak to this from knowledge, but on general principles I would expect that readers do notice any eccentricity in a typeface, whether it's in the direction of more faithfulness to Byzantine models than is usual or more extensive Latinization than usual. And that anything that impairs functionality is not appreciated.

But stylistic Latinization - as opposed to anything that takes liberties with actual letter forms - might well be accepted, at least within limits, warmly - because it makes it easier to have the diversity, the choices, of typefaces that the Latin alphabet has.

I feel that the Byzantine model is inadequate to the typographic needs of the Greek people, and that something that is indigenous to Greece and yet as versatile as the Latin script needs to develop. Latinization is a placeholder until that happens, but it's destructive because it will change what people find natural, leading to future developments being diluted. The alternative, though, is fundamentally unacceptable - people won't do without an effective capability to express themselves in type if one is available.

John Hudson's picture

Tatiana, I do think it is worth noting that the period in which Latinisation seemed to be the only trend in Greek type design -- beginning in the 1970s and extending into the 1990s -- is behind us, and while examples of Latinisation continue to be found, there is a lot more variety of approach and many more designers interested in trying to find an authentically Greek character across a range of styles and typographic needs.

Nick Shinn's picture

Here’s another pertinent essay.
The Greek Futura shown (Olympia, from Karpathakis in 1939) is particularly relevant to the issue of Latinization. There were variants on this theme which further muddied the distinction between Latinization, Modernism, Globalization, and Typeface-ism.

dezcom's picture

As John Hudson said, the problematic period of Latinization is behind us. Today's Greek type designers are free to make Greek typefaces become whatever they see fit. As with any Greek endeavor, there will be differences of opinion on what comes out of it ;-) We all just need to trust in the talent and skill of the young generation of type designers who work in the Hellenic script to proceed in a healthy exploration of their task. We need not restrict their efforts with dictates and decrees of what they MUST do to overcome what their forefathers have left them. What is needed is for us all to just let them design type in their own World and in their own way. They will do what they need to do quite naturally if we don't impede their progress.

Tatiana Marza's picture

Thank you, again, Nick!
This was one of first articles I read. I really don't see the Latinization as ' giving up on history', more as moving forward... That's the problem.

[Edited] Nick,
I read again the article "The loneliness of greek typography; myth or reality". It didn't add anything new to my question regarding the harmfulness of Latinizations, although Klimis Mastoridis displays quite well the reality of Greek typography

John, Nick and Chris, I totally agree with you and I was aware of the fact that Latinization is already behind us. But you see, I began the thesis with the hypothesis that this phenomena implies some consequences and as matter of fact I do not see serious problems. Which would be ok too.

Except of 3 points mentioned by me earlier, can you add some other 'side effects' of Latinization? Much more serious?

Chris, you are a source of wisdom! :)
Thank you for the posted image.

John Hudson's picture

For me, the most obvious effect of Latinisation is not a cultural or ideological one, but a graphic one. When non-Latin scripts are subject to Latinisation, to processes of trying to make them 'more like Latin', whether structurally or in terms of borrowed shapes or details, the result is almost always an awkward stiffness, the loss of organic feel. As I suggested earlier with reference to Gill and Van Krimpen, when Latin designers bring this approach to non-Latin types they seems to demonstrate a loss of the very sensibility that makes them successful designers in the Latin script: they end up producing parodies of their Latin types as much as parodies of the writing system for which they are designing. When native designers bringing a Latinisation approach to their own scripts, I'm often left wondering whether they've ever really looked at the Latin script or tried to understand its subtleties. 'Latin' is often taken as a synonym for simplified, which leads to crude geometries and lack of sensitivity to relative proportions and how letters group together in words.

charles ellertson's picture

You wrote:

Writing the thesis on Latinization of Greek typography I faced a pragmatic question: why is so harmful this phenomena?
. . .
I cannot find a strong argument against Latinization.

I would agree. In a similar vein, I have a friend who is American, ethnically half Cherokee (Cherokee is one of the many American Indian tribes). She feels it very important to preserve the old tribal values, traditions, etc.

I am American, ethnically half Norwegian. I do not feel it important to preserve the old values, traditions, etc., be those 10th century Viking traditions or 19th century Norwegian/American traditions.

* * *

In the States, there is a relatively new argument about the value of 'diversity.' The claim of value usually comes up in the context of race, ethnicity, sexual orientation, etc., and the numbers of people included in/excluded from various groups. But the argument is always put in terms of the value of diversity per se. If that's the case, don't we want to include an ax murderer for maximum value?

The point is, there is usually an agenda in all such claims, charges & countercharges, which if not exactly hidden, is not bought forward, either. Usually because it is spurious.

The knowledge, the history is interesting and has uses. Specific claims to its value are less obvious. You are wise to question them.

Tatiana Marza's picture

BTW (I'm addressing to everyone),

When you design a greek and separately a latin font, which letter you use for your greek x-height and which for your latin x-height? Should it be the letter which has more frequent linguistic occurrence?

In a cross-script font, is there a rule where the greek x-height should be smaller than the latin x-height? Or in every font, according to the final goal, the designer choses the proportions between greek and latin x-height?

Thank you!

Thank you for the encouragement :)

quadibloc's picture

@John Hudson:
For me, the most obvious effect of Latinisation is not a cultural or ideological one, but a graphic one. When non-Latin scripts are subject to Latinisation, to processes of trying to make them 'more like Latin', whether structurally or in terms of borrowed shapes or details, the result is almost always an awkward stiffness, the loss of organic feel.

Yes! And inappropriate or artificial use of historical models can do the same thing.

John Hudson's picture

Charles, a lot of arguments in favour of cultural diversity ultimately reduce to a kind of biological metaphor. We've got excellent reasons to believe that biodiversity is essential for the long-term survival prospects of life on earth, and we presume that cultural diversity might be similarly beneficial for humanity. Personally, I think such arguments are unnecessary, because there are perfectly good short-term reasons to favour cultural diversity over monoculture in terms of richness and variety of experience (which is not the same thing as cultural relativism; being able to say that e.g. circumcision is a form of cultural violence against the individual is important within this diversity) and respect for both particular and shared heritage as something to be celebrated rather than lamented.

That said, in terms of writing systems, and hence typography, a process like Latinisation seems simply perverse to me. Why does this conversation even take place, why is this a phenomenon? We're not debating Chinesification of Danish typography, or Hindification of Malaysian typography, or Cyrillification of Ethiopian typography, because these are all plainly ludicrous. Dressing up one writing system in another's clothes is ludicrous, and one doesn't need to make arguments from cultural diversity to say it is a pathetic idea. As a kind of novelty display typography, it has a long history, but would you as an English reader sit down with a book set in Chinese Takeaway? And yet we consider it normal, or at least something about which reasonable debate is possible, that Greek -- or Thai, or Hindi, or Russian -- readers should read Latinised versions of their writing systems? And that this constitutes 'progress', as if the Latin writing system were in some way inherently superior, and not simply the beneficiary of political, financial and industrial contingency?

Let me also clarify that while I think it unfortunate in terms of loss of richness and variety, I don't blame any society that decides to give up its indigenous writing system in order to adopt one that it thinks will be more materially beneficial. I think the traditional Javanese script is one of the structurally cleverest and most beautiful in the world, but adoption of a Latin orthography in the mid-20th Century clearly helped integrate Java within Indonesia and contributed to increased literacy in the socio-political context of that time. But actually adopting a new script is an entirely different matter than playing dress-up with your old script, of trying to make it look like something else. That's just sad and pathetic.

Now, does this mean limiting the range of expression of your writing system and typography to specific historical models? No, I don't think it does, and I think it is an entirely false association to equate the development of new styles with Latinisation just because Latin type designers have produced such a variety of stylistic innovations over the past couple of centuries. And, indeed, in what way does aping stylistic idioms of Latin typography constitute innovation in another script? If the only way you can stylistically innovate is to try to come up with a Greek -- or Thai, or Hindi, or Cyrillic -- version of Helvetica, then maybe you're writing system is culturally dead on the ground already.

Nick Shinn's picture

I don’t think that’s the issue, John.
It doesn’t matter what the type looks like, people should tell their own stories in their own voices.

A few years ago, western multi-national corporations were producing experimental versions of foreign (to their creative directors and employed/hired type designers) scripts. These are now derided, and such corporations are now politically correct and conservative in respecting the quaintness of the until recently typographically-backward cultures to which they distribute their fonts, advised by expat gatekeepers.

Complicating the issue is the fact that classical Greek is OUR (Western Latin script users) story too, deeply embedded in our academic culture—and this too lends a conservative tone to the culture of Greek type design in the West.

I have nothing against a bit of conservatism, but it seems to me that Greek and Cyrillic type design in the West is overly safe and skewed in that direction.

That is why I look to type designers such as Panos Vassiliou and Alexandra Korolkova for inspiration, and hope that if my efforts at Greek and Cyrillic have a bit of a foreign accent for indigenous readers, it will be sexy, like Garbo or Schwarzenegger.

quadibloc's picture

@John Hudson:
as if the Latin writing system were in some way inherently superior, and not simply the beneficiary of political, financial and industrial contingency?

While I agree this is true, I am not sure that it is always relevant. Cheaper hand-me-down typewriters or computers, an easier time learning economically significant foreign languages - those are as much valid reasons for choosing the Latin script, if making a choice is practical, as any intrinsic superiority.

Of course, there is no choice to be made in the case of Greek, because the people already have a writing system they know how to use, in which many books have been printed. But as a general principle, contingency is as good as inherent superiority for many purposes.

John Hudson's picture

John (Q), note that the fragment of my post that you quoted referred specifically to Latinising indigenous scripts, not to adopting the Latin script. As I said, and think we agree, there are good reasons why someone might decide to abandon one script for another, just as there are good reasons why many people abandon their native language in favour of a dominant regional or trade language. But I don't see any good reason for trying to make your writing system look like someone else's.

quadibloc's picture

@John Hudson:
But I don't see any good reason for trying to make your writing system look like someone else's.
It's true that there are few cases where that would produce obvious benefits.

Still, some writing systems have limited legibility; in Hebrew, some letters are difficult to distinguish, for example.

The fusion of a Roman monumental uppercase with a semi-cursive lowercase in Greek is, I think, a problem - even with the Porson italicization removed. One way to solve it would be to completely Latinize the lowercase, but fortunately that's prevented by the fact that for some of the letters in the alphabet, there's no obvious Roman-style lowercase shape. A typeface that you have to learn how to read has little chance of catching on. So Porson got improved, leading to Apla, and little further movement towards a more Latin-like lowercase for serif fonts has taken hold.

The other way is to go away from the Roman model. The common New Hellenic model goes that way.

John Hudson's picture

Still, some writing systems have limited legibility; in Hebrew, some letters are difficult to distinguish, for example.

Yes, and within the varieties of Hebrew square letter and other styles there are things one can do to reduce these difficulties. The fact that in some of the Ashkenazic styles these things are not done and, if anything, the confusable letters are made more rather than less similar might suggest a number of interesting things about writing, type design and reading.

Mediaeval Arabic writers produced an entire genre of literature based around misreadings of un-disambiguated archigraphemes.

If disambiguation of hard to distinguish characters is the goal, I still don't see any obvious benefit comes from aping Latin letters that could not be attained in other ways. The methods used by cultures that adopted the Arabic script to capture sounds that do not exist in Arabic are a good example of the richness of sufficiently disambiguated forms that can typically be located within a scribal tradition, for instance the adoption of optional variant forms from an Arabic style as normative letters in a new orthography.

Chris Dean's picture

@Tatiana Marza: Be certain to properly cite Typophile in the references section of your paper. A failure to do so would constitute academic plagiarism. APA 6.0 format:

Dean, C. (2013, March 29). Re: Questionnaire about the needs of Greek typography end-users [Online forum comment]. Retrieved from

You will need to do this for every comment you reference.

Is this undergraduate or graduate research?

Tatiana Marza's picture

@Chris Dean
Is this undergraduate or graduate research?

I am an undergraduate student at AKTO Thessaloniki, Greece (, Department of Graphic Design. This research is part of my BA in Graphic Design.

Tatiana Marza's picture

While reading your last comments, I could not ask myself what elements make a Greek alphabet (esp. lowercase letters) look Greek?
The same, of course, could be asked about the Latin alphabet.

| There are some lowercase letters in Latin, which never could be passed for Greek letters and vice versa.
| The greater number of ascenders and descenders in case of Greek letters
| The ductus

During the designing process of a cross font, what other features you are taking in consideration in order to give a Greek appearance to letters?

dezcom's picture

"what elements make a Greek alphabet (esp. lowercase letters) look Greek?"
Ductus is a loaded question. I do not feel that a particular ductus should be a requirement to look Greek overall. Ductus to me can only be addressed if we are looking for some subcategory of Greek that details the writing tool or technique or time period. I should be able to crudely scratch Greek in the dirt with a pointed stick and still have it be Greek.
In other words, if it looks like a duck, walks like a duck, and talks like a duck, it is a duck.

To quote my friend Nick Shinn, what makes it look Greek is the "Curly Bits".
That is an oversimplification but still of value. My mother detested reading what might be called sans serif Greek type (even though there really are no serifs in Greek). She may be referring to the entrance and exit strokes John Hudson referred to or to Nick's curly bits (She died several years ago at 90 so I can't ask.). Greeks of today have differing opinions but are quite accustomed to what we choose to call sans serif.
Arabic and other scripts also have "curly bits" but they do not look Greek. Greek is a phonetic system like Latin and follows the same rules of reading order (at least beyond the earliest form). This may give it a Latin-like grey value on a page from a distance but close up, what distinguishes it from Latin to me is the variety of stroke type. Latin ascenders are quite regular, a straight vertical bar is dominant. Greek has more variety and none of it as simple as a straight line. Descenders in Latin are much more regular and simple as well. The double-boweled g is the only odd-man out. Latin x-height is clearly defined. Greek is much more of a tendency than an exact point. I see Greek more like a form which extends both ways beyond a centerline than Latin, which is pretty much a railroad track.

Greek glyphs borrow less from each-other than Latin (which is much more modular). What I might find a better description is to say that while Latin "marches" [lock-step to the beat of the drum across the page in rigid 4:4 time], Greek "dances" [in complex time signatures to a lilting mix of interplay of instruments].

So then if it looks like a dance, sounds like a dance, and moves like a dance, it must be Greek. Opah!

John Hudson's picture

A word about ductus:

This is a problematic term, because it has a couple of different uses when talking about writing and type. Some of my calligrapher friends are very particular that the word should only be used to refer to the order and direction (and sometimes speed) of strokes by which a letter is written, which is the traditional meaning and provides the etymological base for the word. I've come to the conclusion that they're right, despite having previously used it in the other sense, favoured recently by type designers, to refer to pen angle (and hence to stroke modulation pattern).

Tatiana Marza's picture

Chris, it seems that there is quite enough Greek blood running through your veins;) I am referring to your 'opha!' :)

John, what do you think about the "Greekness" of Greek lowercase letters?

John Hudson's picture

...what do you think about the "Greekness" of Greek lowercase letters?

Isn't that what we were talking about for most of this thread? I don't think I've got much to add other than what I wrote earlier:

I judge Greekness in terms of texture on the page, rather than specific letter features or treatments, but obviously that texture is built up from those features. In other words, I find it much easier to critique Greekness as a quality when looking at a page of text, seeing how the letters work together, than in terms of looking at individual letter shapes.

As someone who works with a lot of different writing systems, I am sensitive to the fact that most scripts evolve normative letter constructions based on particular writing tools and practices, and that typography sits in relation to these letters (it can sit in critical relationship, or in rebellious relationship, but it can't ignore them). In the case of Greek lowercase letters, this means sitting in relationship to Byzantine scribal tools and practices, since that is where these letters came from. This means dealing with a predominantly cursive internal construction.

Does this mean 'Greekness' must involve only reproducing Byzantine forms, or even particular Byzantine cursive script as interpreted by the Aldine press? No. It should also be noted that the Byzantine scribal tradition was not stylistically uniform. The historically most fruitful innovations in Greek typeface design, it seems to me, have engaged with that tradition, looking for ways to reinterpret the normative letter constructions with new tools (Didot), or looking for different directions emerging from that tradition (Scholderer's New Hellenic).

Andreas Stötzner's picture

After all I’m still and again wondering what *Greekness* actually may imply…
For instance, this particular devotion to single letters in their ancient shape, like they practise with the Theta on streets. I can’t help guessing the Greek have a special approach to things which others lack.

dezcom's picture

They didn't have a leg of Lambda to stand on ;-)

Yulia's picture

Tatiana, thank you for starting so interesting research, I'm looking forward to see the results of your study published. Unfortunately I found this thread too late to add any questions for the questionnaire, so I decided just to tell you about making greek script in my Flauto typefamily, maybe it could be of any help. It is a venetian antiqua with faces for different optical sizes and weights and now I have greek in light regular styles (I plan to make greek for bolds and italics too). When I finished first variant of my greek it looked rather latinized so it was hard to distinguish latin and greek in mixed texts. I didn't like it and I decided to make it more traditional so that it could be easily seen (even without reading) as greek (not latin or cyrillic). For the second variant of my greek I've chosen ἁπλά style, because it seems to be closer to venetian antiqua then other traditional greek styles. While making this variant I noticed that greek lowercase seem smaller and lighter then latin and cyrillic ones so I decided to make some compensations: I increased the height of greek lowercase and made them a little darker trying to balance the color and height of all scripts. Here is an example of my greek, latin and cyrillic scripts in a light regular style for the 21-35 pt:

Nick Shinn's picture

Indeed, a nice balance of weight between the scripts.
But no “quaint” ligatures for Greek and Cyrillic?

Yulia's picture

Thank you, Nick. I plan to make “quaint” greek ligatures, especially in italic, and it would be natural for they were used a lot in such style. As for cyrillic ligatures I tried to make them in regular style but they seem a little annoying in text, and decided to leave only ту, ут, тут (which help to improve letter spacing) and several uppercase ligatures. In italic cyrillic ligatures look more natural, so I've already made some and will continue. Today ligatures are still rather rarely seen in cyrillic books in text setting, though some journals (for example «Большой город») use cyrillic typefaces with ligatures, but they are mainly sans serif. So cyrillic ligatures can be seen more often in headlines then in text. Here is an example of my cyrillic ligatures:

Nick Shinn's picture

Given that Venetian Cyrillic is an historical anachronism derived from Latin script, it would be quite logical to also incorporate Latin-style quaints in the Roman, involving the /c, a connecting flourish, and a second letter with an ascender.

That might be ef, or perhaps ka or yu with ascender, Bulgarian-style.

Yulia's picture

Nick, thanks for idea, I think it's worth trying with quaints, especialy for Bulgarian-style cyrillic, which is more cursive in structure and has more letters with ascenders.

Tatiana Marza's picture

Юля, привет вам из Греции! :)

You are more than welcome for the Latinization research! It started out of nothing, as usually things do and had no idea that it would evolve into a captivating part for me, at least.

Yes, is too late for questions, but if you have any, I would be more than happy to include them into my thesis, as a discussion part.

I increased the height of greek lowercase and made them a little darker trying to balance the color and height of all scripts.

The x-height part of letters is preoccupying me intensively these days. I posted more than once this question on this thread, but never got any replies. Having no experience in designing typefaces, could you please, tell me, if the relation of greek x-height to other scripts x-height is always smaller, or it depends on the style of letters a designer choses to give to his font family?

Then, you said, that you made Greek font a little darker. By modifying modulation, stroke width? If you can share with us your secrets, it would be wonderful :)

Your font looks great, BTW! And it happens that I know all three languages, so maybe it's easier for me to extract information. Tell me please, how much you want the 3 scripts to look alike? I am asking because Greek seems to me to be dancing most of all under the music of "flute/flauto", then comes Russian, then Latin.
In Greek language, from all letters, the ''λ" seems a little bit stiff, not 'dancing' so much.
As for the cyrillic italic, the ligatures seem a little bit as ' just inserted there', not emerging from the natural way of writing letters. Seem to be added after you wrote the letters. Maybe you have time to try another kind of ligatures, more 'flaoutique' ? :)

hrant's picture

Yulia, the color in your Greek and Latin are indeed nicely matching. I wonder however whether there's room for improving the harmony in apparent size, apparent leading and spacing.

The x-height part of letters is preoccupying me intensively these days.

Sounds like my life since the late 90s. :-)

I promise I will catch up (at least on that aspect) in the next 2-3 days. And now that I said that in public there's no escape, which is not a bug but a feature. :-)

I know all three languages

This is an enviable position for somebody who wants to design Big Three (Latin, Cyrillic, Greek) typefaces, and if I'm not mistaken, quite rare. It's even better than being left-handed in tennis. :-)


Tatiana Marza's picture

Hrant, you are exaggerating. It's all a matter of luck, in terms of where someone is born. I grew up in former Soviet Union, so I learned Romanian with a cyrillic alphabet, than Romanian with Latin alphabet, Russian was just my second mother language an then I came to Greece, so the languages became a natural advantage of my everyday life ...
Besides the languages I posses, the terrific would be to know Arabic or Hebrew. That would be a true enviable position.

Hrant, you are like Fantomas these period! :)
Yes, please, do come back. We need answers:)

Andreas Stötzner's picture


I’VE GOT IT !!!!

dezcom's picture

Andreas, it just needs a bit of Feta ;-)

Tatiana Marza's picture

Somebody is hungry or is about new forms of Greek alphabet? :)

The secret of typophile Greekness at last: It’s just Paprika.

Amazing gamma! :) No, really!
I'm not sure I ever saw a similar one, referring to the upper part of it.
Will this be your new Cutprika font :), Andreas?

hrant's picture

Tatiana: Yup, you got very lucky! :-)
I have nativity in three scripts too, but: Armenian is very hard to market; Arabic is somewhat hard to market and it's technically difficult.

terrific would be to know Arabic or Hebrew.

But it's too late. After about age 10 nativity is very hard to acquire. The good news is it's still possible to make good type in a script you don't really know, however it's limited to mimicking precedent - it's very hard to innovate in it.

Andreas, that's cool! The garnish is a great touch too.


Andreas Stötzner's picture

> new Cutprika font … ?

Perhaps. :-)
Maybe I go and add Glagolitic.

Out of mushrooms.

Yulia's picture

Татьяна, спасибо!:) The question that interests me is about the using of polythonic small caps inside Greece today. I read John Hudson wrote: “Smallcaps have been used at times to emulate square uncial letters in the transcription of ancient manuscripts”. I've made them in my font, but still wonder If there's a need and does anybody uses it in contemporary editions in Greece.
Answering your question about x-height: my latin and cyrillic scripts have the same x-height, but greek lowercase I've made a little higher, I mean if you compare for example latin “o” and cyrillic “о” with greek “о” (omicron), the omicron will be mathematically higher then latin and cyrillic “o” in order to seem the same optically. I was adjusting the height of greek lowercase till it seemed to me equal to latin and cyrillic in text. And the similiar is about weight: for example if you compare again omicron with latin and cyrillic “o”, the omicron will be darker. I've increased the stroke weight till it seemed to me that greek evened in color with cyrillic and latin in text setting.
And also Yannis Haralambous wrote in his article “From Unicode to Typography, a Case Study: the Greek Script” (a very useful and interesting article that helped me a lot): “An intrinsic characteristic of the ἁπλά family is that the ratio of lowercase height to appercase height is smaller than the one of typical Latin typefaces. This makes harmonization between ἁπλά and Latin typefaces harder.” I've been doing greek in just ἁπλά-style and I can say for my case the compensation of height was needed, but I don't think that it works for all styles or fonts (and there are fonts that don't seem to belong to any particular traditional style), it should be checked by eye every time. However it seems to me the more “curly” or “cursive” (don't know how to formulate the term exactly) the style is, the more it needs to be compensated, but I'm not sure.
And answering your question about my three scripts, while making I tried to harmonise them so that they clearly seemed to belong to one typefamily, but at the same time to keep the tradition of each script and distinguish in mixed texts.
Andreas, what a tasty lettering:)

quadibloc's picture

No doubt Tatiana can tell you what is needed in Greece, but I would like to submit this URL:

of a site which is apparently a voice crying in the wilderness. Essentially, if I correctly remember what I have read on this subject, after the regime of the generals collapsed in Greece, katharevousa, being so strongly identified with them - even if they did not use the best quality of it - collapsed, and dimotiki is used now for all purposes. And this took the polytonic script with it.

I am suspecting, though, that you don't really want to use small caps to simulate ancient inscriptions. Instead, there are special sans-serif typefaces for that purpose; in fact, I thought that one had already been discussed in this thread.

Tatiana Marza's picture

Thank you for your interesting explanation on Greek x-height.
Concerning the usage of "polythonic small caps inside Greece today" I am not able to give you an answer now, because I will violate the integrity of my statistic analysis*. But, as I said, during mid June I will post the results of my questionnaire. Since you have the polythonic small caps, keep them. You never know!

* I may tell you about some clues I have concerning your question, but if in time I receive more answers from the questioners, what I tell you now may be totally irrelevant in 2 months.

Yulia's picture

quadibloc, thank you for the link, I have seen this interesting site, which was organized by Yannis Haralambous. He's doing a fantastic job for spreading polytonic system and involving more people into it. When I started making greek nearly two years ago, first I didn't intend to make polytonic, but after reading Yannis Haralambous I found his arguments so convincing that I decided to make polytonic and alternative forms and some glyphs that didn't have unicode numbers, but were still needed and used in Greece.
Татьяна, I'll wait till the mid June of course, for the integrity of your analysis.

Tatiana Marza's picture

A bit "out of" topic for last comments, but I am trying to analyze the Greek letter (related to what I am writing in my thesis now).
Can anyone explain how exactly the letter frequency in a language can help to understand where the information is distributed (cartesian area) and how this affects the design?

I found the letter frequency for GR alphabet, but how can anyone use this information for designing greek type? Yes, descenders and ascenders hold the last positions and more round shapes took the first places.

α ο τ ε ν ι π ρ σ (ς?) μ υ κ λ η γ δ ω χ θ φ β ξ ζ ψ

α 12
ο 9.8
τ 9.1
ε 8
ν 7.9
ι 7.8
π 5.024
ρ 5.009
σ 4.9
μ 4.4
υ 4.3
κ 4.2
λ 3.3
η 2.9
γ 2.0
δ 1.7
ω 1.6
χ 1.4
θ 1.3
φ 1.2
β 0.8
ξ 0.6
ζ 0.5
ψ 0.2

quadibloc's picture

When I started making greek nearly two years ago, first I didn't intend to make polytonic, but after reading Yannis Haralambous I found his arguments so convincing that I decided to make polytonic and alternative forms and some glyphs that didn't have unicode numbers, but were still needed and used in Greece.

I am glad that you did. I should have noted one thing in my post to avoid the potential of it being misleading. While I believe that modern Greek is not typeset in the polytonic form any more, it is still very important and useful for a Greek font to have polytonic capability (even if extending it to new frontiers like small capitals is dubious) - because most modern Greeks are able to read Classical Greek without having it translated for them, and that is properly typeset with the polytonic accents.

Imagine an English-language font that couldn't be used to print Shakespeare! (But I'm not arguing to have eth and thorn added to every Latin font, nice though that would be. Chaucer, most English speakers do need a translation to read.)

EDIT: I think I may have become misleading in the other direction.

Native speakers of Russian do not need an acute accent indicating the stressed syllable in every word. Native speakers of Hebrew do not even need the vowel points! So native speakers of Greek may indeed have little trouble reading, say, Aristophanes without the use of the polytonic script apparatus.

But a typeface not available in polytonic fonts could not be used for a dictionary of Classical Greek, it couldn't be used for accurate quotations of texts printed in polytonic form, and for so many scholarly purposes, that I think it would be comparable to a Latin font developed by an English speaker that had no accented characters, and so couldn't be used to quote from French or Italian or Spanish - it would be usable only for the simplest purposes.

dezcom's picture

Frequency of letters and combinations can help when kerning is done. As an example, if you find that two glyphs as neighbors require kerning but through frequency, you find that this combination so rarely might occur as to make the need of a kerning pair irrelevant.

John Hudson's picture

As Chris suggests, sequence frequency is more useful than single letter frequency. Pair frequency is useful for kerning; frequency for longer sequences are useful for some writing systems, e.g. consonant conjunct sequences in Indian languages.

I find single letter frequency information most useful in getting a reasonable sense of the comparative fit of different typefaces. So, taking Tatiana's frequency information -- what is the source of this, by the way? -- one can create a frequency-weighted string of text like this:


This can then be used to compare how much space different fonts takes up, and provide a reasonably good guide to their comparative fit. Of course, individual documents might differ significantly from average frequencies, on account of recurring personal names or special terminology. And note that this Greek frequency data does not distinguish regular and final sigma.

hrant's picture

Even better, visualize the differences via frequency.
For English-versus-Armenian:
That's basically what Armenian and English look like, in a single glance.
I've done this for English-versus-Greek too; let me dig it up.


John Hudson's picture

How is that practically useful?

quadibloc's picture

Aside from the usefulness of letter freqency in solving simple substitution cipher puzzles, it gives an idea of the 'feel' of a language. Why does Bodoni suit Italian, Caslon suit English, and Garamond suit French?

Tatiana Marza's picture

@John Hudosn,
Tatiana's frequency information -- what is the source of this, by the way?, look at page 83Ανάλυση_Συχνότητας_Γλώσσας, I found is dated from 2009
But they are in Greek...

And note that this Greek frequency data does not distinguish regular and final sigma.
Yes, this is quite a problem in being totally accurate about letter frequency.

@John Q,
What do you mean by the 'feel' of a language?

I tried to do same kind of interpretation for Greek and Latin alphabet. Although it seems that my information on Latin letter frequency differs from yours...

Syndicate content Syndicate content